JP7803384B2 - Method for manufacturing a negative electrode mixture layer containing graphite particles and sulfide solid electrolyte particles - Google Patents

Description

The present invention relates to a method for producing a negative electrode mixture containing graphite particles and a sulfide solid electrolyte.

In recent years, electronic devices such as mobile phones, tablet devices, and laptop computers have become increasingly sophisticated and functional, leading to growing demand for smaller, lighter, and higher-capacity secondary batteries for these devices. Under these circumstances, non-aqueous electrolyte (liquid) secondary batteries, typified by lithium secondary batteries, are rapidly gaining popularity in the field of these electronic devices because they offer higher battery voltages and higher energy densities than nickel-cadmium batteries and nickel-metal hydride batteries.

Furthermore, against the backdrop of recent environmental concerns, non-aqueous electrolyte secondary batteries have also become mainstream as power sources for driving the motors of electric and hybrid vehicles. However, because non-aqueous electrolyte secondary batteries generally contain flammable organic solvents, they pose safety issues, such as the risk of fire if the secondary battery experiences abnormalities such as thermal runaway, and immediate improvements are needed. As one technology to address these safety issues, active research and development is underway on all-solid-state lithium secondary batteries, which use solid electrolytes instead of organic solvent-based electrolytes.

The all-solid-state lithium secondary battery is mainly composed of a three-layer structure including a cathode composite (layer) containing a cathode active material and a solid electrolyte, a cathode composite (layer) containing a cathode material using In metal, a Li-In alloy, or carbon and a solid electrolyte, and a solid electrolyte (layer) provided between the cathode composite and the anode composite. The solid electrolytes contained in the constituent materials can be broadly divided into oxide-based and sulfide-based solid electrolytes, and typical examples of the _{former oxide - based} solid electrolytes include _{Li7La3Zr2O12 and Li5La3Nb2O12 .}

However, oxide-based solid electrolytes have poor crushability, which can prevent a dense structure from being obtained during the powder compaction process in which powdered raw materials are compressed to form a solid-state lithium secondary battery. Oxide-based solid electrolytes also have issues, such as the need for sintering under high-temperature conditions to achieve high ionic conductivity. In contrast, sulfide-based solid electrolytes do not have the aforementioned issues associated with oxide-based solid electrolytes. In particular, typical sulfide-based solid electrolytes (hereinafter also referred to as LPS) having the Li ₇ PS ₆ phase, Li ₄ P ₂ S ₆ phase, or Li ₃ PS ₄ phase, as disclosed in Patent Document 1, possess the flexibility and adhesiveness characteristic of sulfides, allowing them to be easily deformed during the powder compaction process. Furthermore, sulfide-based solid electrolytes have the advantage that, depending on their composition, high ionic conductivity can be achieved without heat treatment, making them a more promising material than oxide-based solid electrolytes.

The LPS is a compound synthesized using diphosphorus pentasulfide and lithium sulfide as raw materials. Its synthesis method typically involves a reaction synthesis using heat generated by prolonged ball impacts in a planetary ball mill (mechanical milling device) filled with an inert gas, or, if necessary, a heat treatment to precipitate a crystalline phase (metastable phase). Patent Document 2 also discloses a cost-effective technique for producing a negative electrode composite using LPS synthesized by the above method, in which graphite particles, which can insert and remove ions through electrochemical action, are blended as an active material. This technique involves mechanically mixing graphite particles and LPS particles, and then compacting the resulting mixed powder to form the negative electrode composite layer. The LPS contained in the negative electrode composite layer serves as a conduction path for Li ions.

JP 2013-155087 A Japanese Patent Application Laid-Open No. 2014-203545

In the above-mentioned all-solid-state lithium-ion battery (hereinafter also referred to as all-solid-state battery), it is considered preferable to increase the graphite particle content in the negative electrode composite from the perspective of increasing capacity. The reason for this is that a high graphite particle content allows the negative electrode to accept more Li ions that have migrated from the positive electrode. However, when the negative electrode composite is prepared to contain a large amount of graphite particles, depending on the type of graphite particles, it can be difficult to uniformly mix the graphite particles and LPS particles.

In particular, when preparing a negative electrode composite using a dry method, the graphite particles and LPS particles tend to be insufficiently mixed, and the quality of the mixture can affect the quality of the battery characteristics. Therefore, unless the mixture is improved, simply increasing the graphite particle content may not fully utilize the LPS's role as a conductive path, and the desired discharge capacity may not be achieved. The present invention was made in consideration of the above problems, and aims to provide a method for producing a negative electrode composite for all-solid-state batteries that can achieve a high discharge capacity.

In order to achieve the above object, a method for producing a negative electrode composite layer for an all-solid-state battery according to the present invention includes the steps of: weighing out sulfide all-solid-state electrolyte particles having a form of secondary aggregates with an average particle diameter of 60 to 100 μm formed by aggregation of primary particles with an average particle diameter of 0.5 to 2 μm, which are made of an ion-conductive compound composed of lithium, sulfur, and phosphorus; and graphite particles having a particle size distribution D50 of 8 μm to 13 μm , so as to obtain a predetermined blending ratio; dry-mixing the total amount of the weighed graphite particles with an amount of the weighed sulfide all-solid-state electrolyte particles in a range of 20% by mass to 40% by mass for a mixing time of 5 minutes or more; adding the remainder of the weighed sulfide all-solid-state electrolyte particles in an amount in a range of 20% by mass to 40% by mass increments to the dry-mixed mixture, and dry-mixing each amount for a mixing time of 1 minute or less; and compacting the obtained negative electrode composite to form a negative electrode composite layer having a porosity of 3% or less .

The present invention makes it possible to provide an all-solid-state battery that has a high initial charge/discharge capacity and an excellent capacity retention rate in cycle tests.

1 is a photograph of a battery evaluation cell used in an example of the present invention. FIG. 2 is a perspective view of the battery evaluation cell of FIG. 1 with the housing removed. FIG. 2 is a schematic longitudinal cross-sectional view of a battery evaluation cell with the housing removed.

1. Negative Electrode Composite for All-Solid-State Batteries The following describes a negative electrode composite for an all-solid-state battery according to an embodiment of the present invention. This negative electrode composite contains graphite particles and an ionically conductive sulfide solid electrolyte. The negative electrode composite is compacted to obtain a negative electrode composite layer having a porosity of 3% or less. The porosity in the negative electrode composite layer can be determined by the following method. Specifically, a cross section of the compacted negative electrode composite layer is processed using a CP (Cross Section Polisher) device, and five arbitrarily selected fields of view from the mirror-finished cross section are imaged using an SEM. The SEM images of each field are binarized to separate voids from other areas. The ratio of the area occupied by the voids to the total area of the SEM image of each field is determined, and the resulting area ratios are averaged across the five fields. This allows the porosity to be determined.

As described above, by limiting the porosity in the negative electrode composite to 3% or less, it is possible to increase the initial charge/discharge capacity of an all-solid-state battery having the negative electrode composite layer, and also to improve the capacity retention rate in cycle tests. Specifically, during charge/discharge of an all-solid-state battery, Li ions move through the LPS, which acts as an ion conduction path within the negative electrode composite, and thereby insert into and extract from the crystalline layers of the graphite particles. Therefore, as contact between the graphite particles and the surrounding LPS particles decreases, the proportion of voids at their interface (sometimes referred to as the graphite particle/LPS interface) increases. When the porosity in the negative electrode composite layer exceeds 3%, the contact area between the graphite particles and the LPS particles becomes too small, hindering efficient Li ion migration and making it difficult for Li ions to insert into and extract from the graphite particles. In other words, excessive graphite particles that do not contribute to charge/discharge are present in the negative electrode composite layer, significantly reducing the charge/discharge capacity.

As mentioned above, the reason for the formation of voids at the graphite particle/LPS interface is thought to be that the graphite particles and LPS particles are not mixed sufficiently when preparing the negative electrode composite, resulting in insufficient packing of LPS particles around the graphite particles during powder compaction. In other words, LPS particles generally have very poor fluidity, making it difficult to achieve a uniform mixture when mixing graphite particles and LPS particles to prepare the negative electrode composite.

Under these circumstances, we conducted extensive research to minimize the formation of voids at the graphite particle/LPS interface and thereby reduce the porosity in the negative electrode composite layer to 3% or less. We discovered that by using a blend of at least two types of LPS particles—fine particles and particles larger than the fine particles—we can improve the mixability of the graphite and LPS particles and reduce the porosity. Specifically, we found that the porosity can be reduced by blending the LPS particles constituting the negative electrode composite so that, of the total amount of LPS particles, the proportion of fine particles with an average particle diameter of 2 μm to 5 μm is 20% to 40% by mass, and the proportion of large particles with an average particle diameter of 40 μm to 60 μm is 20% to 40% by mass.

As described above, by using LPS particles containing a blend of at least 2-5 μm fine particles and 40-60 μm large particles, good contact between the graphite particles and LPS particles is achieved when compacted, reducing the likelihood of voids forming at the graphite particle/LPS interface, enabling the porosity in the negative electrode composite layer to be 3% or less. This ensures sufficient contact area at the interface between the graphite particles and solid electrolyte particles that make up the negative electrode composite layer, forming numerous ion conduction paths within the negative electrode composite layer and increasing the utilization rate of the graphite particles. As a result, the charge/discharge capacity of the all-solid-state battery can be increased, and the discharge capacity retention rate during cycle testing can also be improved.

In the negative electrode composite of this embodiment, it is preferable to further specify that the graphite particles constituting the negative electrode composite have an average flatness of 0 to less than 0.3, a particle size distribution D50 of 8 to 13 μm, and a BET value of 1 to ^{3 m} /g. By using graphite particles that satisfy these requirements, the contact state between the graphite particles and the LPS particles can be further improved when compacted.

If the average flattening ratio, which is a characteristic related to the shape of graphite particles, is 0.3 or more, the material will contain many irregularly shaped graphite particles, which will hinder the smooth flow of the LPS particles when forming the negative electrode composite layer by powder compaction, making it more likely that voids will form around the graphite particles where the LPS particles are not filled. Conversely, as the average flattening ratio of the graphite particles approaches zero, the shape will change from irregular to spherical, improving the packing of LPS particles around the graphite particles even if the flowability of the LPS particles is poor.

The average flattening ratio of graphite particles can be calculated by the following formula 1. Here, f is the average flattening ratio, a is the major axis of a single particle, and b is the minor axis of a single particle. These radii were calculated by taking images of five arbitrarily selected fields of view on the cross section or surface of a group of graphite particles using an SEM, measuring the major axis and minor axis of five arbitrarily selected graphite particles from each SEM image, and averaging the results.
[Formula 1]
f=1−b/a

If the particle size distribution D50, which is a characteristic related to the average particle size of graphite particles, is less than 8 μm, the graphite particles will be too fine and will be less dispersible when mixed with LPS particles, making it more likely for voids to form around the graphite particles. In other words, if the graphite particles are too fine, they will tend to aggregate, resulting in an insufficient contact area with the LPS particles and an increase in the number of LPS particles that do not conduct ions and graphite particles that do not intercalate or deintercalate Li ions. Conversely, if the particle size distribution D50 exceeds 13 μm, the number of graphite particles per unit thickness of the negative electrode composite layer will decrease, resulting in a decrease in charge/discharge capacity. The particle size distribution D50 of the graphite particles was determined using a laser diffraction method, in which a laser is irradiated onto graphite particles dispersed in a liquid and refracted and scattered light is obtained.

If the BET value, which is a characteristic related to the specific surface area of graphite particles, is less than 1 m ² /g, the graphite particles have less opportunity to come into contact with the LPS particles during mixing to prepare the negative electrode composite, resulting in poor mixing and making it difficult for the LPS particles to adhere to the graphite particle surfaces, making it more likely that voids will form at the graphite particle/LPS interface during powder compaction. Conversely, if the BET value exceeds 3 m ² /g, excessive gas components and moisture will be adsorbed onto the graphite particle surfaces, which can easily cause unstable battery characteristics and make it difficult to obtain high capacity. The BET value can be obtained using the BET formula based on the gas adsorption method (nitrogen adsorption method).

A porosity of 5% or less in the negative electrode composite layer can be achieved simply by specifying the above-mentioned graphite particle properties, such as average flatness, particle size distribution D50, and BET value. However, to further improve the performance of all-solid-state batteries, a porosity of 3% or less is desirable, and it has been difficult to achieve a porosity of 3% or less simply by specifying the above-mentioned graphite particle properties. Although reducing the porosity from 5% to 3% is numerically small, the voids in the negative electrode composite layer are mainly voids that occur at the graphite particle/LPS interface, and the size of the voids is often around 5 to 10 μm. Therefore, reducing the porosity from 5% to 3% has generally been extremely difficult.

For example, even when spherical graphite particles are used, small voids can remain at the interface. Therefore, we conducted extensive research, believing that there must be other factors besides the poor fluidity of the LPS particles mentioned above that could be causing the poor mixability of graphite and LPS particles. As a result, we discovered that when graphite and LPS particles are mixed to produce the negative electrode composite, the LPS sometimes does not adhere to the surface of the graphite particles. In other words, after mixing the graphite and LPS particles, the particles sometimes separate at a microscopic level, and we believe this is the reason why the porosity cannot be reduced.

The inventors therefore conducted further research not only into the properties of graphite particles, but also into LPS, which tends to adhere to the surface of graphite particles. They discovered that when LPS particles are synthesized using the mechanical milling method described below, they are produced in the form of secondary aggregates with particle sizes of approximately 60 to 100 μm, which is the reason for insufficient adhesion to the surface of graphite particles and insufficient mixing with the graphite particles. In this case, it was thought that mixing could be improved by breaking down the LPS particles in the form of secondary aggregates, but LPS secondary aggregates are prone to degradation when mechanically mixed or broken down to obtain single particles; for example, prolonged mixing times could result in a significant decrease in ionic conductivity.

Therefore, we prepared two types of LPS particles with different particle sizes, produced by varying the crushing time, and mixed them with graphite particles to produce a negative electrode composite. This made it possible to reduce the porosity in the negative electrode composite layer to 3% or less. In other words, because the voids that form at the graphite particle/LPS interface are very small in volume, the LPS particles that fill these gaps do not need to be the entire amount of LPS particles used to make the negative electrode composite. Furthermore, in order to adhere to the surface of graphite particles, if the LPS particles adhere along the uneven surfaces formed on the surface of the graphite particles, fine powder LPS particles are preferable, but even in this case, it is sufficient to use an amount that covers one layer of the surface area of the graphite particles.

Therefore, for example, by gradually adding LPS particles to graphite particles during crushing in at least two separate batches, it is possible to mix finer LPS particles obtained by mixing for a long period of time with larger LPS particles that are less susceptible to deterioration obtained by mixing for a short period of time. This makes it possible to produce a negative electrode composite layer that has a low porosity at the graphite particle/LPS interface while minimizing the content of LPS particles damaged by mixing.

Specifically, it is preferable that the total amount of LPS particles used as the negative electrode composite contain 20% to 40% by mass of more highly pulverized fine particles with an average particle diameter of 2 μm to 5 μm, and 20% to 40% by mass of large-diameter particles with an average particle diameter of 40 μm to 60 μm, which are less damaged by mixing. In this way, by using a negative electrode composite in which LPS particles containing predetermined contents of fine particles and large-diameter particles, each having a predetermined particle size range, are blended with graphite particles, the porosity of the negative electrode composite layer after powder compaction can be kept to 3% or less, allowing the production of all-solid-state battery cells with excellent battery characteristics, such as initial charge/discharge capacity.

1. Method for Producing Negative Electrode Mixture Layer Next, an embodiment of the method for producing a negative electrode mixture layer according to the present invention will be described, taking as an example a case where lithium sulfide and diphosphorus pentasulfide are used as raw materials for LPS particles.

1.1 Graphite Particles While natural or artificial graphite can be used as the graphite particles constituting the negative electrode composite, artificial graphite is preferred because it easily satisfies the above-mentioned average flatness, particle size distribution D50, and BET value characteristics. MCMB (Meso-Carbon MicroBeads: spherical carbon particles) are particularly preferred. On the other hand, natural graphite generally has a poor flatness due to its non-spherical shape, a wide particle size distribution D50, and a high BET value, which often results in poor compatibility with LPS particles. Furthermore, amorphous carbon may be provided on the surface of the graphite particles to improve Li ion insertion and extraction during charge and discharge. This facilitates adhesion of LPS particles to the surface of the graphite particles during mixing of the negative electrode composite, enabling a higher charge and discharge capacity to be obtained.

1.2 LPS particles (1) Lithium sulfide (Li ₂ S)
There are no particular restrictions on the lithium sulfide (Li ₂ S) that serves as one of the raw materials for the LPS particles that make up the negative electrode composite, but it is preferable to use an industrially produced commercially available product, and one with a purity of 99% or more is particularly preferred. Lithium sulfide can be produced, for example, by the method described in Japanese Patent No. 3528866. This production method involves blowing hydrogen sulfide into a feed solution containing lithium hydroxide and an aprotic organic solvent placed in a reaction vessel, causing the lithium hydroxide and the hydrogen sulfide to react to produce lithium hydrosulfide, and then subjecting the resulting reaction solution containing lithium hydrosulfide to a dehydrosulfiding treatment to produce lithium sulfide.

(2) Diphosphorus pentasulfide (P ₂ S ₅ )
Although there are no particular restrictions on the phosphorus sulfide that is the other raw material for the LPS particles that make up the negative electrode composite, it is preferable to use commercially available industrially produced diphosphorus pentasulfide (P ₂ S ₅ ). This diphosphorus pentasulfide preferably has a purity of 99% or more.

(3) Blending Ratio When producing a typical high ion conductive sulfide, it is preferable to adjust the blending ratio of lithium sulfide to the total of the lithium sulfide and diphosphorus pentasulfide used as starting materials so that it is within the range of 30 to 95 mol%. The specific blending ratio is determined appropriately depending on the application. For example, as in the case of a 70Li ₂ S-30P ₂ S ₅ solid electrolyte produced using starting materials in which the blending ratio of lithium sulfide and diphosphorus pentasulfide is 70:30 on a molar basis, if the blending ratio of lithium sulfide is 70 mol% or less, the high ion conductive sulfide becomes vitreous and requires heat treatment when producing the high ion conductive sulfide. Furthermore, because it is unstable in the air, handling during synthesis, storage, etc. becomes difficult. Therefore, the blending ratio of lithium sulfide in the starting material for the high ion conductive sulfide, LPS, is preferably within the range of 75 to 80 mol%.

(4) Synthesis of LPS LPS can be obtained by synthesizing the starting materials lithium sulfide and diphosphorus pentasulfide blended in the above-mentioned ratio. This synthesis is preferably performed by mechanical milling. Mechanical milling is a processing method in which a powder to be processed is loaded into a milling container of a mixing device along with a large number of metal or ceramic balls, which act as media that physically act on the powder. The media move in response to the rotation and vibration of the milling container, applying physical forces such as mixing, stirring, and impact to the powder.

The mixing device used in this mechanical milling method is not particularly limited, as long as it can load and mix the above-mentioned media, which uniformly mix and stir the powder to be processed, into a milling vessel. Common powder mixing devices such as planetary ball mills, rotary ball mills, and attritors can be used, but planetary ball mills are particularly preferred. Planetary ball mills use random ball movement, which promotes uniform mixing of the powder to be processed and provides a large amount of reaction energy to the powder, thereby providing a driving force that promotes the synthesis reaction extremely efficiently.

In this way, mechanical milling differs from two-step processing, in which multiple starting material powders are first mixed and then heat-treated, in that heat is generated by the impact of rapidly moving balls, allowing multiple starting material powders to be uniformly mixed and undergo rapid localized chemical reactions simultaneously, making it suitable for producing exceptionally high ionic conductors.

In the mechanical milling method described above, it is preferable to control the atmosphere inside the milling container. For example, when filling a milling container made of stainless steel with _ZrO2 on the inside, it is desirable to create an inert gas atmosphere inside the milling container. Therefore, the milling container is preferably filled in a glove box filled with an inert gas such as Ar gas, _N2 gas, or He gas. However, since the mechanical milling apparatus itself cannot usually be placed inside the glove box, after filling, the milling container is removed from the glove box and milling, such as mixing, is performed in an environment exposed to the outside air. Therefore, the lid of the milling container is preferably designed to hermetically seal the inside of the milling container, preferably with a silicone packing.

The reason why it is preferable to hermetically seal the milling container as described above is that sulfides begin to deteriorate immediately upon exposure to air, and are particularly susceptible to moisture. Even a small amount of moisture in the gas inside the milling container causes decomposition, resulting in the generation of H ₂ S gas. Therefore, it is preferable that the inert gas filling the milling container be sealed in the milling container in a dry state and hermetically sealed with packing as described above to prevent leakage to the outside. Note that gas leakage to the outside can also be considered a state in which moisture-containing air from the outside air enters the milling container. To seal dry inert gas into the milling container as described above, the dew point inside the glove box where the filling operation is performed is preferably controlled to be below −70°C dp, more preferably below −80°C dp (moisture content 0.5 ppm).

1.3 Powder Compaction A negative electrode composite layer for an all-solid-state battery can be produced by compacting the negative electrode composite obtained by mixing the graphite particles and LPS particles. In compacting this negative electrode composite, it is preferable to stack the positive electrode composite, solid electrolyte, and negative electrode composite in this order into three layers and then pressurize them in the stacking direction to simultaneously compact a three-layered green compact. Good compactability is desirable for this negative electrode composite powder compaction. However, because the negative electrode composite of the present invention does not contain a resin binder, the quality of this compactability depends on the adhesiveness of the LPS particles.

To fully utilize the adhesive properties of the LPS particles, it is preferable to mix the graphite particles with the LPS particles in a ratio of 50 to 70 parts by mass per 100 parts by mass of the total of the graphite particles and LPS particles. This allows the graphite particles and LPS particles to be compacted into a compact with excellent shape retention and no defects or cracks.

In contrast, if the graphite particles are blended in an amount exceeding 70 parts by mass per 100 parts by mass of the graphite particles and LPS particles combined, the shape retention may be reduced, resulting in cracks. Such cracks in the compact may increase resistance and reduce charge/discharge capacity. Conversely, if the blending ratio of graphite particles is less than 50 parts by mass per 100 parts by mass of the graphite particles and LPS particles combined, the carbon content derived from the graphite particles in the negative electrode composite layer obtained after compaction is too low, making it unable to sufficiently accept Li ions transferred from the positive electrode active material, again resulting in reduced charge/discharge capacity. One possible solution to this problem would be to compensate for the carbon content by increasing the thickness of the negative electrode composite layer, but this is not a desirable solution because increasing the thickness of the negative electrode composite layer increases resistance accordingly.

As mentioned above, LPS particles immediately after synthesis by mechanical milling preferably have the form of secondary aggregates (secondary particles) with an average particle size of 60 to 100 μm, formed by the aggregation of primary particles with an average particle size of 0.5 to 2 μm. Dry mixing is preferably used to mix these secondary aggregates of LPS particles with graphite particles. This is because wet mixing can result in a violent reaction between the solvent used for mixing and the LPS, which can cause deterioration, whereas dry mixing does not present such a problem. Furthermore, since synthesized LPS particles are vulnerable to friction and impact during mixing, and MCMB can also be crushed or distorted when force is applied, which can result in a deterioration of battery performance, a mixing method that minimizes damage to the LPS and graphite particles is preferred. Therefore, hand mixing using a mortar, as described below, is preferred. However, when using mechanical milling, such as a planetary ball mill, as in the production of LPS particles, mixing at the minimum rotation speed and for the shortest possible time is desirable.

A more preferable mixing method, in terms of minimizing the damage mentioned above, is the mixing method using a mortar. By appropriately adjusting the force applied to the powder during mixing and the mixing time, mixing can be achieved with little damage. However, when LPS particles and graphite particles, weighed out in the desired ratio, are added to a mortar and pestle and mixed for 10 minutes or more, the LPS particles are easily crushed into a fine powder, but are damaged by this crushing. When the LPS particles are damaged in this way, the adhesive properties inherent to the sulfide in LPS are reduced, making it difficult to function as a binder. This can lead to the formation of many voids at the interface between the graphite and LPS particles when the negative electrode composite layer is formed. Furthermore, the ionic conductivity of LPS itself is reduced, which can increase the resistance of the negative electrode composite layer fabricated using it and reduce battery capacity. Furthermore, when LPS particles are crushed into fine particles, their activity increases, and they are subject to deterioration due to trace amounts of moisture and oxygen in the environment, reducing their shelf life and making them unsuitable for storage.

Therefore, in an embodiment of the present invention, LPS particles and graphite particles are mixed using a mortar in the following manner. That is, first, the LPS particles and graphite particles are weighed out to achieve the desired blend ratio. Next, the total amount of the weighed graphite particles and 20% to 40% by mass of the total amount of the weighed LPS particles are charged into the mortar, and a first mixing is performed using a pestle at a speed of approximately 30 to 120 revolutions per minute, taking care not to press the mixture, so as to achieve a uniform overall mixture. Secondary agglomerates of LPS particles are easily broken down by mixing with the pestle for approximately 5 minutes, yielding primary particles and secondary agglomerates (secondary particles) with particle sizes of approximately 2 μm to 5 μm. Therefore, this first mixing is performed for 5 minutes.

This makes it easier for the LPS particles to adhere to the surface of the graphite particles, allowing the surfaces to be covered with LPS particles. Although the LPS particles are slightly damaged during this first mixing and crushing process, the benefit of reduced porosity at the graphite particle/LPS interface is greater than this disadvantage, as it is more effective in increasing capacity. However, by minimizing the ratio of fine particles to the total amount of LPS particles weighed above, the adverse effects of the damage are limited to only those that fill the voids.

After the first mixing, all of the remaining weighed LPS particles are added in amounts ranging from 20% to 40% by mass to the mixture in the mortar after the first mixing. Each time an addition is made, mixing is performed with a pestle for one minute using the same mixing method as above. In this way, the LPS particles added during the second and subsequent mixings are less likely to be damaged due to the short mixing time, and therefore the LPS particles added during the second and subsequent mixings have almost no adverse effect on the characteristics of the negative electrode composite layer. However, because the mixing time is short, disintegration cannot be carried out sufficiently, and as a result, the particle size of the LPS particles is approximately 40 μm to 60 μm, with secondary agglomerates only slightly disintegrated.

In other words, in the negative electrode composite manufacturing method according to an embodiment of the present invention, a first 5-minute mixing step forms fine LPS particles with a particle diameter of 2 μm to 5 μm, which then coat the surfaces of the graphite particles. Then, each subsequent 1-minute mixing step mixes relatively large LPS particles with a particle diameter of 40 μm to 60 μm, which are less damaged, with the graphite particles whose surfaces are coated with the fine LPS particles. This improves the filling of voids at the graphite particle/LPS interface with the fine LPS particles, and the relatively large LPS particles maintain their adhesiveness, resulting in a synergistic effect that reduces the porosity within the negative electrode composite layer formed by the powder compaction to 3% or less. As a result, a high-capacity all-solid-state battery can be fabricated.

The average particle size (also referred to as average aggregate size) of the secondary aggregates of the above-mentioned LPS particles was determined by taking images of five randomly selected fields of view using an SEM, measuring the outer diameters of five randomly selected secondary aggregates from each SEM image, and averaging them. This is because the laser diffraction method, which involves irradiating a conventional laser and obtaining refracted and scattered light, cannot be used because most of the water or organic solvents used chemically react with LPS and are altered. It is also possible to calculate the particle size from the BET value, but this has problems such as the LPS deteriorating during handling.

When evaluating the charge/discharge capacity of an all-solid-state battery fabricated by stacking the above-mentioned negative electrode composite with a solid electrolyte made of LPS particles used in the negative electrode composite and a separately prepared positive electrode composite, the negative electrode composite, positive electrode composite, and solid electrolyte samples for evaluation are compacted and measured. The thickness of the compacts compacted from each sample is determined taking into consideration the balance between shape retention during compaction (no breaks or cracks during compaction) and resistance, with the positive electrode composite layer and negative electrode composite layer each having a thickness of approximately 100 to 200 μm, and the solid electrolyte layer having a thickness of approximately 300 to 500 μm.

When forming a negative electrode composite layer from the above-mentioned negative electrode composite, for example, the above-mentioned negative electrode composite is filled into a mold with an inner diameter of 10 mm and pressed in the thickness direction at a pressure of approximately 2 to 4 kN to perform temporary powder compaction, followed by high-pressure powder compaction. During this high-pressure powder compaction, the positive electrode composite, solid electrolyte, and negative electrode composite are stacked from the bottom up, and a pressure of 40 kN is applied in this stacking direction to simultaneously compact the three layers. This results in a three-layer integrally structured compact. An all-solid-state battery can be manufactured by incorporating the three-layer integrally structured compact thus formed into a commercially available battery cell case (sealed type: not exposed to the atmosphere). Next, the present invention will be described in more detail based on the following examples and comparative examples, but the present invention is not limited to these examples.

[Example 1]
(Synthesis of LPS)
A total of 80 g of _ZrO2 balls (ball diameter 4 mm) were placed in a milling container (volume ₄₅ ml; inner wall made of ZrO2) attached to a Fritsch mechanical milling device (PL-7) and dried in vacuum at 100°C. This dried milling container was placed in a glove box (-80°C dp) in an Ar atmosphere and further dried overnight.

A milling vessel in the glove box was charged with 0.3828 g of lithium sulfide (Li ₂ S) manufactured by Junsei Chemical Co., Ltd. and 0.6172 g of diphosphorus pentasulfide (P ₂ S ₅ ) manufactured by Sigma-Aldrich and sealed. In this case, the molar ratio of lithium sulfide to diphosphorus pentasulfide was Li ₂ S:P ₂ S ₅ = 75:25. After sealing, the milling vessel was removed from the glove box and placed in the mechanical milling apparatus (PL-7) manufactured by Fritsch GmbH installed in a dry room (ambient temperature 22°C: -45°C dp), whereupon mechanical milling was carried out at 510 rpm for 20 hours. After completion of mechanical milling, the milling vessel was returned to the glove box, the lid of the milling vessel was opened, and the LPS in the form of powder produced by synthesis was removed and passed through a 100-mesh sieve.

The obtained under-sieve LPS particles were analyzed using a non-exposed Raman spectrometer (Microscopic Laser Raman Spectrometer LabRAM HR-800, Light Source: Ar+ Laser (514 nm Excitation), High Excitation Output: 0.2 mW, Measurement Range: 100-4000 cm ^-1 ) manufactured by Horiba, Ltd., and it was confirmed that they contained a Li ₃ PS ₄ phase. Furthermore, to confirm the degree of disintegration when the LPS particles synthesized by the mechanical milling method were mixed in a mortar, 120 mg of the under-sieve LPS particles were sampled and divided into six samples. Five of these samples were separately placed in a mortar and disintegrated in the mortar for 1 minute, 3 minutes, 5 minutes, 10 minutes, and 20 minutes, respectively. The average value of five agglomerate diameters (secondary particle diameters) in the SEM images was then calculated for each of the six samples, including these five samples and one undisintegrated sample.

As a result, the uncrushed sample in the as-synthesized state had an average particle size of 89 μm, the sample crushed for 1 minute had an average particle size of 53 μm, the sample crushed for 3 minutes had an average particle size of 19 μm, and the sample crushed for 5 minutes had an average particle size of 3 μm. Meanwhile, the sample crushed for 10 minutes had re-aggregated the finely pulverized powder, reaching an average particle size of 21 μm. The sample crushed for 20 minutes had re-aggregated in the same way as the sample crushed for 10 minutes, and had an average particle size of 26 μm.

(Preparation of negative electrode composite)
80 mg of graphite particles (PG11A) manufactured by Rongtan Electronic Materials Co., Ltd., with an average particle flatness of 0.03, a particle size distribution D50 of 10.2 μm, and a BET value of 1.2 m ² /g, were added to a mortar in an Ar atmosphere glove box (-80°C dp). 20 mg of the LPS particles that had been sieved were then added. The first mixing was performed for 5 minutes at a rotation speed of 60 revolutions per minute using a pestle, taking care not to press the particles. Next, 20 mg of the LPS particles that had been sieved were added to the mixture in the mortar after the first mixing, and a second mixing was performed for 1 minute using the same mixing method using a mortar. The addition of 20 mg of LPS particles and mixing for 1 minute were repeated two more times to perform the third and fourth mixings. This produced a powder-like negative electrode composite as an intermediate product, in which graphite particles and LPS particles were blended in a 50:50 ratio by mass.

(Preparation of positive electrode composite)
A lithium-nickel composite oxide powder prepared in advance using a known technique was used as the cathode active material. Specifically, nickel oxide powder primarily composed of Ni was mixed with lithium hydroxide and calcined to produce a lithium-nickel composite oxide powder (hereinafter _referred to as NCA particles) serving as the cathode active material expressed as _{Li1.090Ni0.76Co0.14Al0.10O2 .} The average particle size of the resulting NCA particles was measured using a laser diffraction method and found to be _{10.6 μm} . The specific surface area was also measured using the BET method and found to be 0.16 ^m2 /g. 120 mg of the NCA particles and 80 mg of the undersized LPS particles were placed in a mortar in an Ar atmosphere at -80°C dp in a glove box. The mixture was gently mixed with a pestle at a rotation speed of 60 revolutions per minute for 10 minutes, taking care not to press the particles together.

(Preparation of powder compact)
In a glove box (-80°C dp) with an Ar atmosphere, 60 mg of the above-sieved LPS powder was placed in a cylindrical mold with an inner diameter of 10 mm and pre-molded at 2 kN to prepare a solid electrolyte layer. 15 mg of the above-prepared granular negative electrode composite was stacked on top of this solid electrolyte layer and pre-molded again at 2 kN. The obtained two-layer compact was turned over, and 20 mg of the above-prepared positive electrode composite powder was deposited on both sides of the solid electrolyte layer opposite the pre-molded negative electrode composite layer side, followed by final molding at 40 kN. This produced a three-layer integrated compact consisting of a positive electrode composite layer / solid electrolyte layer (SE layer) / negative electrode composite layer. This compact was placed in a sealed battery evaluation cell (KP-solidCell) manufactured by Hohsen Co., Ltd. as shown in Figures 1 and 2, and the evaluation cell was assembled by restraining it with a torque of 7 Nm. As shown in FIG. 3 , this evaluation cell has an integrally structured green compact 10 made up of a negative electrode composite layer 11, a solid electrolyte layer (SE layer) 12, and a positive electrode composite layer 13, which can be pressed in the stacking direction with a predetermined restraining force by an upper electrode 21 and a lower electrode 22 within a cylindrical insulating tube 20.

(Evaluation of Battery Characteristics and Measurement of Porosity of Negative Electrode Mixture Layer)
The terminals of the obtained evaluation cell were connected to a charge/discharge device (HJ-SD8) manufactured by Hokuto Denko Corporation, and the initial discharge capacity was determined over a voltage range of 2.37 to 4.17 V_CC (0.1 C = 19 mAh/g), resulting in a value of 124 mAh/g. Furthermore, the discharge capacity retention rate in a cycle test was determined by repeating charge/discharge 20 times under the same conditions as the initial charge/discharge, and examining the change in capacity between the initial discharge and the 20th cycle. This result was 92%. After evaluating the battery characteristics, the porosity of the negative electrode composite layer was determined as an average of five SEM images as described above, resulting in a value of 2.4%.

[Example 2]
An intermediate negative electrode composite containing graphite particles and LPS particles in a 50:50 mass ratio was prepared in the same manner as in Example 1, except that 30 mg of LPS particles were added instead of 20 mg in the first and third mixing runs and the fourth mixing run was omitted. This intermediate negative electrode composite was used to prepare a three-layer integrated compact consisting of a positive electrode composite layer, a solid electrolyte layer (SE layer), and a negative electrode composite layer in the same manner as in Example 1. This compact was then placed in a sealed battery cell in the same manner as in Example 1 to assemble an evaluation cell. The battery characteristics were evaluated and the porosity of the negative electrode composite layer was measured in the same manner as in Example 1. The initial discharge capacity was 123 mAh/g, the discharge capacity retention rate was 91%, and the porosity of the negative electrode composite layer was 2.6%.

[Example 3]
An intermediate negative electrode composite was prepared in the same manner as in Example 1, except that graphite particles (PW8A) manufactured by Rongtan Electronic Materials Co., Ltd., with an average flatness of 0.24, a particle size distribution D50 of 8.7 μm, and ^a BET value of 1.2 m/g, were used. This intermediate negative electrode composite was then used to prepare a three-layer integrated compact consisting of a positive electrode composite layer, a solid electrolyte layer (SE layer), and a negative electrode composite layer, in the same manner as in Example 1. This compact was then placed in a sealed battery cell to assemble an evaluation cell, and the battery characteristics were evaluated and the porosity of the negative electrode composite layer was measured in the same manner as in Example 1. The initial discharge capacity was 120 mAh/g, the discharge capacity retention rate was 90%, and the porosity of the negative electrode composite layer was 2.9%.

[Comparative Example 1]
An intermediate negative electrode composite in which the graphite particles and LPS particles were blended in a 50:50 ratio by mass was prepared in ^the same manner as in Example 1, except that 80 mg of graphite particles (AC1) manufactured by Rongtan Electronic Materials Co., Ltd. (having an average flatness of 0.42, a particle size distribution D50 of 5.3 μm, and a BET value of 2.4 m/g) and 80 mg of the undersized LPS particles used in Example 1 were placed in a mortar and mixed once for 10 minutes using a pestle in the same manner as in Example 1. This intermediate negative electrode composite was used to prepare a three-layer integrated powder compact consisting of a positive electrode composite layer/solid electrolyte layer (SE layer)/negative electrode composite layer in the same manner as in Example 1. This powder compact was then placed in a sealed battery cell in the same manner as in Example 1 to assemble an evaluation cell, and the battery characteristics were evaluated and the porosity of the negative electrode composite layer was measured in the same manner as in Example 1. As a result, the initial discharge capacity was 104 mAh/g, the discharge capacity retention rate was 79%, and the porosity of the negative electrode mixture layer was 6.7%.

[Comparative Example 2]
An intermediate negative electrode composite was prepared in the same manner as in Comparative Example 1, except that graphite particles (OMAC-R) manufactured by Osaka Gas Chemicals, having an average flatness of 0.37, a particle size distribution D50 of 11.8 μm, and a BET value of 4.4 m ² /g, were used instead of the graphite particles (AC1). The graphite particles and LPS particles were blended in a 50:50 ratio by mass. This intermediate negative electrode composite was then used to prepare a compact having a three-layer integral structure consisting of a positive electrode composite layer, a solid electrolyte layer (SE layer), and a negative electrode composite layer in the same manner as in Example 1. Furthermore, this compact was placed in a sealed battery cell in the same manner as in Example 1 to assemble an evaluation cell. The battery characteristics were then evaluated and the porosity of the negative electrode composite layer was measured in the same manner as in Example 1. The results were an initial discharge capacity of 105 mAh/g, a discharge capacity retention rate of 80%, and a porosity of the negative electrode composite layer of 5.3%. The measurement results for Examples 1 to 3 and Comparative Examples 1 and 2 are summarized in Table 1 below.

10 Green compact 11 Negative electrode composite layer 12 Solid electrolyte layer (SE layer)
13 Positive electrode composite layer 20 Insulating tube 21 Upper electrode 22 Lower electrode

Claims (1)

A method for producing a negative electrode composite layer, comprising the steps of: weighing out sulfide all-solid-state electrolyte particles having a form of secondary aggregates with an average particle size of 60 to 100 μm formed by aggregation of primary particles with an average particle size of 0.5 to 2 μm, the sulfide all-solid-state electrolyte particles being made of an ion-conductive compound composed of lithium, sulfur, and phosphorus; and graphite particles having a particle size distribution D50 of 8 μm to 13 μm, so as to obtain a predetermined blending ratio; dry-mixing the total amount of the weighed graphite particles with an amount of the weighed sulfide all-solid-state electrolyte particles in a range of 20% by mass to 40% by mass for a mixing time of 5 minutes or more; adding the remainder of the weighed sulfide all-solid-state electrolyte particles in an amount in a range of 20% by mass to 40% by mass increments to the dry-mixed mixture, and dry-mixing each amount for a mixing time of 1 minute or less; and compacting the obtained negative electrode composite to form a negative electrode composite layer with a porosity of 3% or less .